INDEX

Explanations

words related to legal actions and procedures

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sinclair

-0.65

hander

-0.62

fer

-0.59

 Koran

-0.59

ciating

-0.58

grown

-0.58

sted

-0.58

Graphics

-0.57

fortable

-0.57

 Feld

-0.57

POSITIVE LOGITS

uary

1.40

uate

1.25

uated

1.22

inic

1.07

uating

1.07

ional

0.99

uation

0.98

ivism

0.96

ual

0.96

ively

0.93

Activations Density 0.591%

words related to legal actions and procedures

No Comments

No Known Activations